Amomum tsao-ko —Chinese Black Cardamom: Detailed Oil Composition and Comparison With Two Other Cardamom Species

Abstract

Two Chinese black cardamom oils (Amomum tsao-ko) were analyzed using GC and GC/MS and compared to 2 Indian cardamom oils from the species Amomum subulatum and Elettaria cardamomum, respectively. The main constituents of A. tsao-ko oils were eucalyptol, geranial, geraniol, trans-2,3,3A,7A-tetrahydro-1H-indene-4-carbaldehyde, (2E)-decenal, neral, and 4-indanecarbaldehyde. Special attention was given to the 1H-indene-carbaldehyes, which are frequently wrongly assigned in previous literature. A detailed odor evaluation of the oils was also carried out. In addition, composition variations of 28 main constituents of 8 E. cardamomum oils were investigated, taking various factors like origin, production methods, plant material, and drying stage into consideration.

Spices ¹ and the resulting essential oils and oleoresins ² are key components in flavor creations. Some spices, like cardamom, also play an important role in the fragrance industry. ³ Cardamom essential oil for fragrances is hydrodistilled from seeds or less frequently from whole pods of Elettaria cardamomum (L.) Maton. ^4,5 However, the term cardamom is not limited to only one species. Several plants in the genera Elettaria, Amomum, and Aframomum, all belonging to the Zingiberaceae family, are also referred to as cardamom. ¹ In the Encyclopedia of Herbs and Spices, 8 different cardamoms ¹ are described in detail while ISO 676:1995 (Spices and Condiments—Botanical Nomenclature) lists 10 species. ⁶ Our Scent Expeditions to Kerala/India in 2007 and Yunnan/China in 2011, respectively, sparked our fascination for cardamom further.

Cardamom is often distinguished into true or green cardamom on one hand represented by the species E. cardamomum and false or black cardamom referring oftentimes to a variety of species. ⁵ The color classification is based on their dried fruit appearance. Further indexing of cardamom refers to the fruit size (eg, small or large), form (eg, round), or the historic region of origin like China and Madagascar. ¹

Green, small, or true cardamom, E. cardamomum, is known as the “Queen of Spices” ^4,5,7-10 . When bleached it is named as white cardamom. ^8,9 Its essential oil is one of the oldest raw materials in perfumery and the third most expensive spice after vanilla and saffron. It is cultivated widely in the tropical region with Guatemala, India, and Sri Lanka as the main producer countries currently. ^4,5,7-10 The biggest growing regions in India are Kerala followed by Karnataka. ¹⁰ At least 2 varieties of E. cardamomum are listed in ISO 676:1995: var. major Thwaites and var. minuscula Burkill. ⁶ Elettaria cardamomum var. minuscula could be better classified into 3 cultivars according to literature: Malabar, Mysore, and Vazhukka, the latter a hybrid between Malabar and Mysore. ^11,12 Besides the essential oil, CO₂ extracts, as well as the oleoresin, are commercially available. These products are normally obtained only from the seeds by removing the outer husk before processing.

Black, large, or false cardamom covers the species Amomum and Aframomum. ¹³ By far the most important species is Amomum subulatum Roxb., also known as Nepal cardamom native to the Eastern Himalayan region. ^{14 -17} India is the main producer followed by Nepal and Bhutan. In India, 85% of A. subulatum is cultivated in the Indian state of Sikkim but also partly the Darjeeling district in West Bengal, Assam, and Himachal Pradesh. ¹⁴ While large quantities of the spice itself are commercially available, the respective essential oil and oleoresin are hardly found.

The same applies for its Chinese relative Amomum tsao-ko Crevost & Lemarié. ^13,18 However, production volumes seem to be significantly lower and the capsule fruit is bigger compared to A. subulatum. ¹⁹ It is an important spice in many parts of China and also used in traditional Chinese medicine. ¹⁸ Chinese black or tsao-ko cardamom is widely distributed in the southwestern provinces of China like Yunnan and Guangxi, but also in the northern highlands of Vietnam and Laos. ²⁰ In China 90% of the crop has its origin in Yunnan province.

While green cardamom (E. cardamomum) ^7-12,21-23 and black cardamom (A. subulatum) from India ^14-17,24-31 are scientifically well investigated over the years, significantly less is published about Chinese black cardamom (A. tsao-ko). ^18,32-36 In addition, most publications have been published only in Chinese ^{37 –48} and some compounds have been repeatedly wrongly assigned.

Here we report detailed analyses of 2 A. tsao-ko essential oils distilled from crushed whole pods originated from China and compared with 2 other cardamom essential oils from the species A. subulatum and E. cardamomum, respectively, both from India in terms of constituents and odor. In addition, E. cardamomum essential oils distilled from dried and fresh pods from India were compared with 3 commercial oils from India and Guatemala and a CO₂ extract from whole dried pods regarding main constituents (>0.2%) to investigate variations in this species.

All 4 cardamom essential oils were hydrodistilled under standard lab conditions using ground dried pods (for details see Experimental section). Yields ranged from 0.7% to 1.8% for A. tsao-ko and 0.9% to 1.5% for A. subulatum and up to 5.2% for E. cardamomum. The 4 cardamom oils A-D were analyzed by GC and GC/MS (Table 1). Constituents were identified by comparing their mass spectra with known compounds, published spectra, or the Symrise in-house library. ⁴⁹ In all 4 cardamom oils A-D, a broad range of major, minor, and trace compounds were identified. Compounds >0.1% are listed in Table 1. These constituents represent a total of 91.6% (A), 91.8% (B), 94.0% (C), and 97.7% (D), respectively. All 4 cardamom oils contained as a common feature high amounts (>20%) of eucalyptol (=1,8-cineol). High eucalyptol levels are not limited to cardamom oil but are also found in eucalyptus, rosemary, laurel leaf, Spanish spike lavender, and Spanish sage oils. ³

Table 1

Chemical Composition (Area %) of Main Constituents (>0.1%) of 4 Cardamom Oils. ^a

No	RI	Compound	A	B	C	D	No	RI	Compound	A	B	C	D
1	1031	α-Pinene	2.3	1.1	4.3	1.3	41	1704	α-Terpineol	3.5	3.3	8.2	1.4
2	1122	β-Pinene	2.1	1.2	6.3	0.3	42	1707	α-Terpinyl acetate	-	-	-	44.8
3	1129	Sabinene	0.2	0.2	0.9	3.1	43	1709	Borneol	0.2	0.1	0.2	<0.1
4	1158	δ-3-Carene	0.3	0.1	t	t	44	1712	Methyl (2E)-decenoate	-	-	-	0.1
5	1165	Myrcene	0.3	0.3	1.3	1.7	45	1721	Germacrene D	<0.1	0.1	0.7	0.1
6	1175	α-Phellandrene	3.0	2.6	<0.1	<0.1	46	1733	β-Selinene	-	-	0.4	0.4
7	1190	α-Terpinene	0.2	0.1	0.6	0.5	47	1737	Geranial	1.6	7.8	-	0.3
8	1210	Limonene	2.7	1.7	6.9	2.9	48	1739	(2E)-Decenyl acetate	1.5	0.7	-	-
9	1226	Eucalyptol	28.1	22.6	52.8	27.9	49	1741	(5Z)-Dodecenal	-	-	-	0.1
10	1254	γ-Terpinene	0.6	0.7	1.3	0.9	50	1745	Bicyclogermacrene	-	-	0.4	-
11	1279	p-Cymene	3.3	0.7	0.3	<0.1	51	1758	Geranyl acetate	3.1	2.7	t	1.2
12	1294	Terpinolene	0.2	0.1	0.3	0.4	52	1763	α-Terpinyl propionate	-	-	-	0.1
13	1295	1-Octanal	0.4	0.4	-	0.2	53	1765	δ-Cadinene	0.2	0.2	0.2	-
14	1309	4,8-Dimethyl-(1,3E,7)-nonatriene	t	t	-	0.1	54	1766	Citronellol	0.1	0.3	-	-
15	1333	2-Hexylfuran	0.6	0.2	-	-	55	1770	γ-Cadinene	-	-	0.1	0.1
16	1342	6-Methyl-hept-5-en-2-one	0.1	0.2	-	<0.1	56	1799	Myrtenol	<0.1	<0.1	0.1	-
17	1436	(2E)-Octenal	0.4	0.7	-	<0.1	57	1812	(Z)-Sabinol	0.9	0.2	-	-
18	1470	trans-Sabinene hydrate	<0.1	-	-	0.2	58	1818	(2E)-Decenol	0.9	0.3	-	-
19	1477	1-Octyl acetate	0.1	0.1	-	0.1	59	1849	Geraniol	7.1	7.0	-	2.2
20	1504	1-Decanal	<0.1	0.1	-	<0.1	60	1852	trans-2,3,3A,7A-Tetrahydro-1H-indene-4-carbaldehyde	3.7	6.8	-	-
21	1536	(2E)-Octenyl acetate	<0.1	0.2	-	t	61	1865	(2E)-Dodecenal	0.9	2.5	-	-
22	1538	2-Octylfuran	0.3	0.1	-	-	62	1883	cis-2,3,3A,7A-Tetrahydro-1H-indene-4-carbaldehyde	1.9	2.3	-	-
23	1545	Linalool	0.6	0.6	0.1	1.1	63	1924	2,3,3A,7A-Tetrahydro-1H-indene-5-carbaldehyde b	<0.1	0.1	-	-
24	1556	cis-Sabinene hydrate	-	<0.1	t	0.2	64	1944	(2E)-Dodecenyl acetate	0.6	0.3	-	-
25	1557	(3E)-Decenal	-	0.1	-	-	65	1972	2,3,3A,7A-Tetrahydro-1H-indene-5-carbaldehyde b	0.2	0.4	-	-
26	1558	Linalyl acetate	-	-	-	0.7	66	1989	4-Indanecarbaldehyde	4.3	2.3	-	-
27	1559	1-Octanol	0.2	-	-	0.1	67	2025	(2E)-Dodecenol	0.4	0.1	-	-
28	1570	trans-p-Menth-2-en-1-ol	0.1	0.1	0.1	0.2	68	2041	(E)-Nerolidol	3.6	2.6	2.1	0.7
29	1574	(E)-Isocitral	<0.1	0.2	-	-	69	2049	5-Indanecarbaldehyde	1.7	2.1	-	-
30	1612	4-Terpinenol	1.3	1.2	3.2	3.0	70	2083	Elemol b	0.4	0.3	-	-
31	1615	(2E)-Octenol	<0.1	0.1	-	-	71	2130	Spathulenol	0.1	0.1	0.6	-
32	1616	(2Z)-Decenal	<0.1	0.1	-	-	72	2193	T-Muurolol	0.1	-	0.1	-
33	1633	cis-p-Menth-2-en-1-ol	0.1	0.1	0.1	0.1	73	2214	Carvacrol	0.4	0.1	-	-
34	1650	(2E)-Decenal	3.0	6.1	-	<0.1	74	2230	α-Eudesmol	0.2	0.1	-	-
35	1659	δ-Terpinyl acetate	-	-	-	0.3	75	2234	Isospathulenol	-	-	0.1	-
36	1664	(E)-Pinocarveol	0.1	0.1	0.1	-	76	2240	β-Eudesmol	0.3	0.2	-	-
37	1678	δ-Terpineol	0.3	0.4	1.0	0.1	77	2241	α-Cadinol	-	-	0.2	-
38	1688	Neral	1.0	4.7	t	0.2	78	2299	Tricosane	-	-	0.1	-
39	1697	p-Mentha-6,8-dien-3-ol	0.2	0.3	-	-	79	2352	2,6-Dimethoxy-4-methyl-phenol	0.1	<0.1	t	-
40	1699	γ-Muurolene	0.2	0.2	0.2	-	80	2358	(2E,6E)-Farnesol	0.1	0.2	-	0.1

RI: retention index relative to C7-C30 n-alkanes on DB-Wax column; t: trace (≤0.01%); -: not detected; A/B: A. tsao-ko, C : A. subulatum, and D: E. cardamomum;

^aMeasured on DB-Wax column;

^bcorrect isomer not determined.

In both A. tsao-ko oils (A/ B) similar major and minor compounds were detected but some showed significant variations in concentration. The composition of the 2 A. tsao-ko essential oils (A/B) revealed, besides eucalyptol (28.1%/22.6%), the following main ingredients (compounds >1.5%): geranial (1.6%/7.8%), geraniol (7.1%/7.0%), trans-2,3,3A,7A-tetrahydro-1H-indene-4-carbaldehyde (3.7%/6.8%), (2E)-decenal (3.0%/6.1%), neral (1.0%/4.7%), 4-indanecarbaldehyde (4.3%/2.3%), (E)-nerolidol (3.6%/2.6%), α-terpineol (3.5%/3.3%), p-cymene (3.3%/0.7%), geranyl acetate (3.1%/2.7%), α-phellandrene (3.0%/2.6%), limonene (2.7%/1.7%), (2E)-dodecenal (0.9%/2.5%), cis-2,3,3A,7A-tetrahydro-1H-indene-4-carbaldehyde (1.9%/2.3%), α-pinene (2.3%/1.1%), 5-indanecarbaldehyde (1.7%/2.1%), β-pinene (2.1%/1.2%), and (2E)-decenyl acetate (1.5%/0.7%). This is well in line with some previously published analyses. ^{18,20,31,33-35,39,45} However, one previous analysis reported very high levels of limonene together with other major compounds which were not present in our analysis. ²⁵

Both oils contained very characteristic indane (=2,3-dihydro-1H-indene) derivatives: 4-indanecarbaldehyde (1) and 5-indanecarbaldehyde (2) and the respective dihydroindane compounds trans- and cis-2,3,3A,7A-tetrahydro-1H-indene-4-carbaldehyde (3/4) (Figure 1).

Figure 1

Chemical structure of 1H-indene-carbaldehyes 1 to 6 from Amomum tsao-ko.

The indanecarbaldehydes 1 and 2 have been isolated and characterized in 1978 from A. tsao-ko. ⁵⁰ The paper is published under the species name Amomum medium Loureiro, which is synonymous with A. tsao-ko. ^50,51 Nevertheless, especially compound 2 has been wrongly assigned in several publications as α- or β-methyl cinnamic aldehyde, ^20,40 2-phenyl-2-butenal, ³⁶ and 3-methyl-3-phenyl-2-propenal, ³⁵ having the same molecular mass (M = 146) as 1 and 2. Also our MS-database showed α-methyl cinnamic aldehyde as a possible hit for 2. Re-isolation of aldehydes 1 and 2 confirmed the structural assignment published in literature (see also the “Experimental” section) with 3 distinct aromatic H-atom signals in ¹H-NMR for both 1 and 2. ⁵⁰ The wrong assignment is probably linked to the different mass spectra of 1 and 2, especially the significantly higher [M-H]⁺ peak of 2 compared with 1. While m/z = 145 is the base peak of (E)-α- and (E)-β-methyl cinnamic aldehyde, both 1 and 2 have m/z = 119 as the base peak. The mass spectrum of 2-phenyl-2-butenal is very close to aldehyde 1; however, the base peak is m/z = 115 compared to m/z = 117 in 1 and 2. Aldehydes 1 and 2 were most likely also present in A. tsao-ko oil from Vietnam, but were wrongly assigned here, as well as α-methyl cinnamic aldehyde. ⁵²

The respective dihydroindane-4-carboxylaldehydes 3 (trans) and 4 (cis) have also been detected in an A. tsao-ko extract and synthesized by Starkenmann et al. ⁵³ Also compounds 3 and 4 have been wrongly characterized as p-propyl ^32,35 or isopropyl benzaldehyde ^34-37,41 in previous A. tsao-ko analyses, all having a molecular mass of M = 148, but mass spectra are far off from compounds 3 and 4. The cis/trans-ratio can vary depending on the interconversion of 3 into 4, as according to Starkenmann the cis-isomer 4 is more stable compared with the trans-isomer 3. ⁵³ Oil A showed a trans/cis-ratio of 1.96:1 while B had a ratio of 2.92:1. Oil B had a higher amount of 3 and 4, but a lower amount of 1 compared to oil A.

Interestingly, to the best of our knowledge, the respective dihydroindane-5-carbaldehydes have not been described before in A. tsao-ko. Two unknown compounds showed high similarities in the mass spectra with 3 and 4. These compounds represent most likely the respective tetrahydro-1H-indene-5-carbaldehydes 5 and 6. However, the characterization is only based on mass spectra similarities and needs further verification by isolation and full spectroscopic characterization as only comparison of mass spectra can lead as shown above to wrong structural assignments.

The indane derivatives 1 to 6 are characteristic for A. tsao-ko and also comprise the structurally related compounds tsaokoin ⁵⁴ and isotsaokoin ⁵⁵ isolated from a CHCl₃ and a methanol extract, respectively. None of these indane derivatives 1 to 6 are present in the other 2 cardamom oils C and D from the species A. subulatum and E. cardamomum, respectively. A literature search confirmed that compounds 1 to 4 have to the best of our knowledge so far not been found in any other species. Therefore aldehydes 1 to 4 can be used as a species marker for A. tsao-ko probably independent of the growing region as they were found in oils from both China and Vietnam. ⁵²

High levels of straight chain unsaturated aldehyde (2E)-decenal together with the presence of (2E)-decenol and (2E)-dodecenol and the corresponding acetates seem to differentiate A. tsao-ko further from the other 2 cardamom species.

The concentrations of eucalyptol, p-cymene, and indane-4-carbaldehyde (1) were lower while those of citral (geranial/neral), (2E)- decanal, (2E)-dodecenal, and trans-tetrahydroindene-4-carbaldehyde 3 were significantly higher in oil B compared with oil A. Citral variations described in literature before for A. tsao-ko were linked not only to fruit forms and therefore to genic variations, ²⁰ but also to different growing regions. ⁴⁰ Higher citral levels were also found in seeds compared with husk, ⁴⁰ but as whole fruits were processed this factor is not relevant here. Fruits distilled for oil A showed indeed a more conical shape while pods used for B had an elliptic form. Composition differences of natural products can be caused not only by genetic factors ²⁰ but by various factors like geographical origin, ⁴⁰ harvest time, date and year, as well as abiotic (eg, climate/weather, soil) and biotic factors (such as herbivores, parasites). However, we are unable to rule out any of these factors as both batches of pods were commercially purchased from different sources and the authors rely on the information obtained by the suppliers.

A detailed analysis of additional minor compounds (<0.1%) from A. tsao-ko is shown in Table 2. These 88 constituents represent only 1.7% to 2.0% of the total oil content and were detected in most cases in both oils A and B.

Table 2

Chemical Composition (Area %) of Minor Constituents (<0.1%) of 2 Amomum tsao-ko Oils. ^a

No	RI	Compound	A	B	No	RI	Compound	A	B
1	1069	α-Fenchene	<0.05	<0.05	45	1692	p-Menth-6-en-2-one	<0.1	<0.1
2	1080	Camphene	<0.1	<0.05	46	1698	(E)-Methyl geraniate	t	<0.05
3	1086	Hexanal	<0.05	<0.05	47	1707	Geranyl formate	<0.05	<0.05
4	1134	δ-2-Carene	<0.05	t	48	1728	Neryl acetate	<0.05	-
5	1141	p-Mentha-1(7),8-diene	t	-	49	1731	p-Mentha-1,5-dien-8-ol	<0.05	<0.1
6	1180	Pentanal	<0.1	<0.1	50	1733	α-Muurolene	<0.1	<0.1
7	1190	2,3-Dehydro-1,8-cineol	<0.05	<0.05	51	1735	1,8-Epoxy-p-menthan-2-ol	<0.1	-
8	1203	1,8-Epoxy-2-p-menthene	<0.05	<0.05	52	1745	Carvone	<0.05	<0.05
9	1236	(Z)-β-Ocimene	<0.1	<0.1	53	1750	(E)-Piperitol	<0.1	<0.1
10	1249	(E)-β-Ocimene	-	t	54	1785	p-Mentha-1(7),2-dien-8-ol	<0.05	<0.1
11	1261	p-Mentha-1(7),2,8-triene	t	t	55	1789	Cuminaldehyde	<0.05	t
12	1276	m-Cymene	<0.05	t	56	1792	1,4-Cadinadiene b	<0.05	<0.1
13	1291	2-Octanone	t	-	57	1804	Nerol	<0.1	<0.1
14	1318	2-Heptanol	<0.05	t	58	1820	Geranyl propionate	-	<0.05
15	1337	2,8-Epoxy-p-menth-6-ene	t	t	59	1829	(E)-Anethol	<0.05	<0.1
16	1351	1-Hexanol	t	t	60	1838	trans-Carveol	<0.1	<0.05
17	1394	2-Nonanone	<0.1	<0.05	61	1841	(E)-Calamenene	t	t
18	1400	1-Nonanal	<0.1	<0.1	62	1859	Guaiacol	<0.05	<0.1
19	1403	Rose furan	<0.1	<0.1	63	1894	epi-Cubebol	<0.05	<0.05
20	1408	(2Z)-Octenal	t	<0.05	64	1947	Cubebol	<0.1	<0.1
21	1421	Perillene	t	t	65	1955	4-Methyl guaiacol	<0.05	<0.05
22	1447	1-Octen-3-ol	t	t	66	2000	o-Cresol	<0.05	<0.05
23	1455	1-Heptanol	<0.05	t	67	2003	Phenol	-	<0.05
24	1468	α-Cubebene	<0.05	t	68	2032	4-Ethyl guaiacol	<0.1	<0.1
25	1483	Citronellal	t	<0.05	69	2065	Cubenol	-	<0.1
26	1500	α-Campholenic aldehyde	<0.05	t	70	2066	1,10-di-epi-Cubenol	<0.1	<0.1
27	1507	α-Copaene	<0.1	<0.1	71	2074	epi-Cubenol	<0.1	-
28	1518	2-Nonanol	<0.05	<0.05	72	2079	2,4-Dimethyl-phenol	<0.05	<0.05
29	1542	(2E)-Nonenal	t	t	73	2090	p-Cresol	<0.05	<0.05
30	1549	β-Cubebene	<0.05	<0.05	74	2094	Guaiol	<0.05	<0.05
31	1556	p-Menth-5-en-2-one	<0.05	<0.05	75	2104	Cuminic alcohol	<0.05	t
32	1557	2-Acetyl furan	<0.05	-	76	2108	Dihydroeugenol	-	<0.05
33	1581	Isopulegol	-	<0.1	77	2167	Eugenol	<0.05	<0.05
34	1585	Pinocarvone	<0.05	<0.1	78	2176	γ-Eudesmol	<0.1	<0.1
35	1589	Fenchol	<0.05	<0.1	79	2176	4-Ethyl-phenol	<0.1	<0.1
36	1592	Bornyl acetate	<0.1	<0.1	80	2185	Thymol	<0.05	<0.05
37	1599	Methyl caprinate	t	-	81	2190	Geranyl heptanoate	<0.05	-
38	1603	Undecan-2-one	<0.05	<0.05	82	2203	δ-Cadinol	<0.1	<0.1
39	1608	Rose furan epoxide	<0.1	<0.1	83	2217	Bulnesol	<0.05	<0.05
40	1633	trans-p-Mentha-2,8-dien-1-ol	-	<0.1	84	2222	3,4-Dimethyl-phenol	<0.05	-
41	1641	Myrtenal	<0.05	<0.1	85	2290	Geranyl caprylate	<0.05	-
42	1660	Alloaromadendrene	<0.1	<0.1	86	2352	2,6-Dimethoxy-4-methyl-phenol	<0.05	<0.05
43	1676	cis-p-Mentha-2,8-dien-1-ol	t	t	87	2411	4-Ethyl-2,6-dimethoxy-phenol	<0.05	<0.05
44	1683	α-Humulene	<0.05	<0.05	88	2482	2,6-Dimethoxy-4-propyl-phenol	<0.05	<0.05

RI: retention index relative to C7-C30 n-alkanes on DB-Wax column; t: trace (≤0.01%); -: not detected;

^aMeasured on DB-Wax column;

^bcorrect isomer not determined.

Amomum subulatum oil (C) showed the highest amount of eucalyptol (52.8%) of all 3 cardamom oils A/B-D. The following other main compounds (>1.5%) were detected: α-terpineol (8.2%), limonene (6.9%), β-pinene (6.3%), α-pinene (4.3%), 4-terpinenol (3.2%), and (E)-nerolidol (2.1%). This is in line with previously published analyses. ^24,56 Eucalyptol levels can go up to 86% ²⁹ and some analyses do not show limonene at all or at very low levels. ³⁰ This is not linked to the fact that whole pods and not just seeds were distilled as the husk does not contribute much to limonene levels. ⁵⁷

Other than in the 2 Amomum species subulatum and tsao-ko which did not contain α-terpinyl acetate, this was the major ingredient in E. cardamomum oil D. The essential oil from dried pods D contained 44.8% α-terpinyl acetate followed by eucalyptol (27.9%), sabinene (3.1%), 4-terpinenol (3.0%), limonene (2.9%), geraniol (2.2%), and myrcene (1.7%). It is interesting to mention that besides α-, also δ-terpinyl acetate (0.3%) was detected. This compound was identified in laurel leaf oil, which also contained the combination of α-terpinyl acetate and eucalyptol, but in lower and opposite ratio. ⁵⁸ The ratio of α-terpinyl acetate and eucalyptol can vary significantly depending on the variety of E. cardamomum. ⁹

A detailed odor evaluation by our perfumers of the neat oils A-D on smelling strips is given in Table 3. We used E. cardamomum oil D from whole pods as reference.

Table 3

Odor Description of 5 Cardamom Oils. ^a

Species	Odor description
Amomum tsao-ko (A)	Strong smoky-leathery, sweet dill, sharp aldehydic-nitrilic, not reminiscent of D; dry-down: smoky, cinnamic, aldehydic-nitrilic, fresher compared to C
A. tsao-ko (B)	Strong smoky-cresolic, more rounded, less leathery and aldehydic-nitrilic compared to A; dry-down: same as A but more rounded and less aldehydic-nitrilic
A. subulatum (C)	Strong smoky-phenolic, birch tar like, aromatic herbaceous, camphoraceous with aspects reminiscent of D; dry-down: strong smoky-tobacco, herbaceous, reminiscent of extracted wet coffee powder
Elettaria cardamomum (D) dried	Warm spicy, fresh aromatic, slightly herbaceous, camphoraceous, eucalyptus; dry-down: warm spicy, floral, fresh-green peely citrus with gingery aspects
E. cardamomum (E) fresh	Similar to D but much more rounded and less eucalyptus

^aOn smelling strips as neat oil.

The oils from the 3 cardamom species gave a very characteristic odor profile. Although the A. tsao-ko oils (A) and (B) were slightly different from an olfactive point they could be clearly distinguished from oils C and D. Especially the smoky notes of both Amomum species A-C are interesting to mention as they were completely absent in green cardamom oil D. These smoky notes can be sometimes attributed to simple phenol derivatives. Besides carvacrol and thymol in oil A and small amounts of eugenol and derivatives a variety of simple phenolic compounds (Table 4) are indeed present in both A. tsao-ko oils (A/B) as well as in A. subulatum oil (C) but not in E. cardamomum oil (D).

Table 4

Phenolic Compounds of 3 Amomum Oils. ^a

No	RI	Compounds	A	B	C
1	1859	Guaiacol	<0.05	<0.1	<0.05
2	1955	4-Methyl guaiacol	<0.05	<0.05	<0.05
3	2000	o-Cresol	<0.05	<0.05	0.1
4	2003	Phenol	-	<0.05	<0.1
5	2014	Eugenol methyl ether	-	-	t
6	2032	4-Ethyl guaiacol	<0.1	<0.1	<0.1
7	2079	2,4-Dimethyl-phenol	<0.05	<0.05	<0.05
8	2082	2-Ethyl-phenol	-	-	<0.05
9	2090	p-Cresol	<0.05	<0.05	<0.1
10	2108	Dihydroeugenol	-	<0.05	<0.05
11	2167	Eugenol	<0.05	<0.05	<0.05
12	2176	4-Ethyl-phenol	<0.1	<0.1	<0.1
13	2185	Thymol	<0.05	<0.05	-
14	2214	Carvacrol	0.4	0.1	-
15	2222	3,4-Dimethyl-phenol	<0.05	-	-
16	2263	3-Ethyl-5-methyl-phenol	-	-	<0.05
17	2352	2,6-Dimethoxy-4-methyl-phenol	0.1	<0.1	t
18	2411	4-Ethyl-2,6-dimethoxy-phenol	<0.05	<0.05	-
19	2482	2,6-Dimethoxy-4-propyl-phenol	<0.05	<0.05	-

RI: retention index relative to C7-C30 n-alkanes on DB-Wax column; t: trace (≤0.01%); -: not detected; A/B: A. tsao-ko, C : A. subulatum

^aMeasured on DB-Wax column.

Table 5

Chemical Composition (Mean Area %) of Major Constituents (>0.2%) of 6 Elettaria cardamomum Oils. ^a

No	RI	Compound	SD	D	E	F	G	H	I
1	1031	α-Pinene	0.11	1.3	1.2	1.9	1.9	1.6	1.0
2	1122	β-Pinene	0.02	0.3	0.5	0.4	0.4	0.3	0.3
3	1129	Sabinene	0.20	3.1	4.1	3.3	3.2	3.6	2.8
4	1165	Myrcene	0.27	1.7	1.9	2.1	1.9	1.7	1.2
5	1190	α-Terpinene	0.15	0.5	0.3	0.5	0.4	0.2	t
6	1210	Limonene	0.20	2.9	2.3	2.9	2.9	2.3	1.8
7	1226	Eucalyptol	0.33	27.9	20.9	24.7	25.4	22.6	24.6
8	1254	γ-Terpinene	0.22	0.9	0.6	0.9	0.8	0.4	0.1
9	1279	p-Cymene	0.22	0.1	t	0.3	0.5	0.2	0.1
10	1294	Terpinolene	0.09	0.4	0.5	0.4	0.3	0.1	<0.1
11	1470	trans-Sabinene hydrate	0.01	0.2	0.5	0.3	0.3	0.5	0.4
12	1474	trans-Sabinyl acetate	<0.01	<0.1	t	<0.1	<0.1	0.6	1.8
13	1477	1-Octylacetate	<0.01	0.1	0.2	0.2	0.1	0.1	<0.1
14	1545	Linalool	0.03	1.1	0.9	0.9	0.6	3.9	0.3
15	1556	cis-Sabinene hydrate	<0.01	0.2	0.4	0.2	0.2	0.3	0.3
16	1558	Linalyl acetate	0.04	0.7	0.5	1.6	1.3	6.5	1.3
17	1612	4-Terpinenol	0.03	3.0	2.0	2.0	1.9	1.1	0.6
18	1659	δ-Terpinyl acetate	0.01	0.3	0.1	0.3	0.3	0.2	0.3
19	1688	Neral	0.01	0.2	0.1	0.3	0.2	0.4	0.3
20	1704	α-Terpineol	0.17	1.4	3.1	0.4	0.4	0.7	0.1
21	1707	α-Terpinyl acetate	0.81	44.8	48.7	46.6	47.4	44.1	53.1
22	1733	β-Selinene	0.02	0.4	0.1	0.8	1.1	0.4	1.6
23	1737	Geranial	0.05	0.3	0.3	0.4	0.1	0.5	0.3
24	1758	Geranyl acetate	0.03	1.2	2.3	1.5	1.3	0.8	0.5
25	1770	γ-Cadinene	<0.01	0.1	<0.1	0.2	0.2	0.2	0.2
26	1849	Geraniol	0.04	2.2	2.1	1.3	1.1	1.3	0.8
27	2041	(E)-Nerolidol	0.05	0.7	0.7	1.2	1.2	1.4	1.1
28	2358	(2E,6E)-Farnesol	0.01	0.1	0.3	0.2	0.1	0.1	0.1

RI: retention index relative to C7-C30 n-alkanes on DB-Wax column; SD: max. ± standard deviation; t: trace (≤0.01%); -: not detected; D: dried pods India, E: fresh pods India, F/G: seeds—commercial India, H: seeds—commercial Guatemala, I: dried pods India—CO₂ extract.

^aMeasured in triplicate on DB-Wax column.

Such compounds can be either natural products or linked to smoky wood fires. ⁵⁹ Zerihun et al describe that different wood fires gave very characteristic phenolic patterns. ⁵⁹ Drying or curing in wood-fired houses is at least described for A. subulatum ¹⁷ and also for E. cardamomum. ^10,60 As the same compounds were detected in both Amomum species from India and China the likelihood that all 3 samples have been dried using smoke from the same wood is minimal and therefore these compounds should be most likely considered part of the natural minor compounds spectrum in these 2 species. However, further investigation is needed to prove this theory. It is worth mentioning that most phenolic compounds described here were found in various smoke flavors before ⁶¹ but also in birch tar oil, ⁶² which might be the reason for the odor reminiscence. In addition phenolic compounds are linked to antioxidant properties, which were described before in literature for both Amomum species subulatum ⁶³ and tsao-ko. ^32,64

Nature sometimes harbors odor secrets even in well-known products, and smelling fresh green E. cardamomum pods during our Scent Expedition to Kerala in India was one of these moments for our perfumers. Fresh green cardamom pods are normally harvested 3 to 5 months after flowering and converted by drying into the final spice product. ¹¹ The odor of the oil from fresh green pods E was much rounder with less eucalyptol smell compared to the oil from dried pods D (Table 3). The odor differences were even more significant to commercial seed oils. However, the oil yield (<2.1%) from fresh green undried pods was also significantly lower compared with the oil from dry pods (<5.2%). To investigate this from a composition point of view we used our steam-distilled oil from dried pods D as a reference and compared it with an oil from fresh green pods E, 3 commercial steam-distilled seed oils of Indian F/G and Guatemalan H origin, and a CO₂ extract from dried pods I (Table 5). This exercise also helped to understand the composition variability of E. cardamomum oils based on various factors like origin (India/Guatemala), production methods [steam distillation (lab/industrial scale)/CO₂ extraction], plant material (seeds/pods), and drying stage (fresh/dried). This picture is still simplistic as it leaves harvest time ¹¹ and annual variations due to abiotic and biotic environmental factors out of the equation.

The 28 compounds listed in Table 5 represent between 94.6% (E) and 96.1% (D) of the total green cardamom oil. Cardamom oil D was used as a reference for comparison. Only compounds having an absolute composition difference (>0.7%) are listed in the text below to reflect differences between the oils. The small sample size does not allow a general statement but reflects only a variation snapshot.

The oil from fresh green cardamom pods E showed significantly lower eucalyptol and 4-terpinenol levels and higher amounts of α-terpinyl acetate, α-terpineol, geranyl acetate, and sabinene compared with D. ^11,65 These composition differences can explain the odor differences. The 2 Indian steam-distilled commercial oils from seeds F/G were very close in composition to each other with only minor differences in the 2 major ingredients. The analyses of E. cardamomum oils (F/G) are well in agreement with previous analyses. ⁶⁶ The oils F/G, however, showed some differences to the lab-distilled oil D, which used whole pods and not just seeds. The commercial oils F/G contained lower amounts of eucalyptol, 4-terpinenol, α-terpineol, and geraniol and higher amounts of α-terpinyl acetate and linalyl acetate. The variances to the commercial oil from Guatemala H were even more significant ⁶⁶ : eucalyptol, 4-terpinenol, geraniol, α-terpineol, and α-terpinyl acetate levels were lower but linalyl acetate, linalool, and (E)-nerolidol content was significantly higher in comparison to lab oil D from India. This picture does not change much by looking only at the commercial Indian seed oils (F/G) vs the Guatemalan seed oil (H): α-terpinyl acetate, eucalyptol, and 4-terpinenol amounts were lower and linalyl acetate and linalool content was significantly higher in the oil H from Guatemala. High levels of linalyl acetate are also found in oils from Sri Lanka ⁶⁶ and Costa Rica ⁶⁷ and recently in a study from Turkey, ⁶⁸ but it stays unclear if the cardamom really originated from Turkey or was only purchased there. Literature shows that linalool and linalyl acetate levels may depend significantly more on the variety and genotype cultivated than on the country of origin. ¹¹

The CO₂ extract I from whole pods gave lower amounts of eucalyptol, 4-terpinenol, geraniol, α-terpineol, limonene, linalool, geranyl acetate, and γ-terpinene but higher amounts of α-terpinyl acetate, trans-sabinyl acetate, and β-selinene in comparison to oil D. This picture changed slightly when the commercial Indian seeds oils F/G were compared to the CO₂ extract from whole pods I. Nevertheless here also significant differences were found despite the fact that some other compounds contributed to this: lower 4-terpinenol, limonene, geranyl acetate, α-pinene, myrcene, and γ-terpinene amounts were found in I but higher amounts of α-terpinyl acetate, trans-sabinyl acetate, and β-selinene compared to oils F/G. Our results are in contrast to previous investigations ⁶⁹ where nearly no differences were found between a steam-distilled oil and a CO₂ extract.

From this small study of green cardamom it can be clearly seen that many factors contribute to variations in natural products, however, to a different extent.

In summary, despite the fact that most essential oils are well investigated they still hold many secrets worthy of proper analysis and discovery. Many places in the world still harbor real treasures ⁷⁰ and some Asian spices are definitively belonging in this category. ^1,71 These spices, for example, both Amomum species, have not even reached the global flavor palette, ³ not to mention the fragrance industry, despite the fact that they are used as culinary ingredients in large quantities in their respective countries of origin. ¹

In addition composition variations in natural products like those shown here for E. cardamomum and A. tsao-ko need to be seen more as fact than as burden. This is a neglected topic in a world targeting not only standardized synthetic aroma molecules but also standardized natural products like essential oils. If the trend for more naturalness continues as anticipated and consumers aim for real authentic natural products, education of customers and consumers is required to make them aware that such variations are a given in any natural product. ⁷² We should not forget that the chance for long-term survival of plants and other living organisms depends on such variations and is a prerequisite for the adaptability of a species to the environment and consecutively allows natural selection which can be the origin of new varieties and species. ⁷³

Experimental

General

GC/MS: Agilent 6890N GC (DB-Wax: 60 m × 0.25 mm × 0.25 μm film thickness, carrier gas He, 60°C-240°C at 3°C/min) connected to Agilent 5975B quadrupole mass spectrometer, 70 eV (EI mode), mass range 25 to 550 amu. GC: Hewlett Packard 6890 with FID and sniffing port (for column and temperature program, see GC/MS). NMR spectra (δ/ppm, J/Hz) were recorded on a Bruker Avance III with 600 MHz (¹H-NMR) or 151 MHz (¹³C-NMR) in C₆D₆ and using TMS as internal standard. All solvents and reagents are commercial products (Merck) and were used as received.

Plant Material

Dried pods of A. tsao-ko were purchased at a spice market in Shanghai, China, and a traditional Chinese medicine retail store in Singapore. Both samples originated from Yunnan province in China according to the suppliers. Dried A. subulatum as well as dried and fresh E. cardamomum pods were obtained from Synthite Industries Ltd, Kolenchery, Kerala, India. Black Nepal cardamom A. subulatum had its origin in the North Eastern Indian state of Sikkim, while the green cardamom was harvested in Kerala state in Southern India. Fresh green pods were collected in the Idukki district of Kerala at an age of 3 months after flowering. All 3 commercial cardamom essential oils from India and Guatemala were sourced in India and France from different suppliers on the open market. The CO₂ extract was received from Ultra International Ltd, Ghaziabad, Uttar Pradesh, India.

Oil Isolation

Whole fresh and dried cardamom pods (100 g) were crushed by hand in a mortar and further ground in a commercial fruit blender followed by hydrodistillation in a Clevenger-type circulatory distillation apparatus ⁷⁴ for 4 hours to yield the respective essential oils. Amomum tsao-ko A/B: yellow oil (0.7%-1.8%); A. subulatum C: yellow oil (0.9%-1.5%), E. cardamomum: dried—pods D: colorless to slightly yellow oil (up to 5.2%), and fresh pods E: colorless to slightly yellow oil (0.6%-2.1%).

Spectral Data for 1H-Indene-Carbaldehydes 1-6

(a) 2,3-Dihydro-1H-indene-4-carbaldehyde (=4-indanecarbaldehyde) (1) ⁵⁰ : RI_DB-WAX 1989. GC/MS m/z (%): 39 (7), 51 (6), 63 (11), 65 (7), 89 (9), 91 (20), 115 (64), 116 (26), 117 (100), 118 (15), 128 (11), 131 (8), 145 (25), 146 (83) [M]⁺, 147 (9). ¹H-NMR: 1.70 (tt, J = 7.6 Hz, 2H, 2H), 2.49 (m, J = 7.6 Hz, 2H, 1H), 3.06 (m, J = 7.6 Hz, 2H, 3H), 6.95 (dd, J = 7.6 Hz, 1H, 6H), 7.03 (m, J = 0.8, 7.6 Hz, 1H, 7H), 7.34 (m, J = 0.8, 7.6 Hz, 1H, 5H), 9.93 (s, 1H, CHO). ¹³C-NMR: 25.31 (t, C-2), 31.89 (t, C-3), 32.06 (t, C-1), 126.76 (d, C-6), 129.41 (d, C-5), 129.53 (d, C-7), 132.96 (s, C-4), 146.13 and 146.17 (s, C-3a, C-7a), 191.82 (d, CHO).

(b) 2,3-Dihydro-1H-indene-5-carbaldehyde (=5-indanecarbaldehyde) (2) ⁵⁰ : RI_DB-WAX 2049. GC/MS m/z (%): 39 (8), 51 (6), 63 (11), 65 (8), 89 (9), 91 (23), 115 (69), 116 (18), 117 (100), 118 (12), 145 (86), 146 (89) [M]⁺, 147 (9). ¹H-NMR: 1.65 (tt, J = 7.5 Hz, 2H, 3H), 2.48 to 2.51 (m, J = 7.5 Hz, 4H, 1H, and 3H), 6.93 (m, J = 7.7 Hz, 1H, 7H), 7.44 (m, J = 0.5, 7.7 Hz, 1H, 6H), 7.50 (m, J = 0.5 Hz, 1H, 4H), 9.78 (s, 1H, CHO). ¹³C-NMR: 25.43 (t, C-2), 32.33 (t, C-3), 33.08 (t, C-1), 124.81 (d, C-7), 125.33 (d, C-4), 128.84 (d, C-6), 136.10 (s, C-5), 145.13 (s, C-3a), 151.31 (s, C-7a), 191.27 (d, CHO).

(c) trans-2,3,3A,7A-Tetrahydro-1H-indene-4-carbaldehyde (3) ⁵³ : RI_DB-WAX 1852. GC/MS m/z (%): 39 (11), 41 (8), 51 (8), 65 (11), 77 (12), 78 (7), 79 (9), 91 (100), 92 (19), 105 (15), 115 (14), 117 (14), 119 (37), 120 (12), 129 (7), 130 (9), 133 (6), 147 (8), 148 (40) [M]⁺.

(d) cis-2,3,3A,7A-Tetrahydro-1H-indene-4-carbaldehyde (4) ⁵³ : RI_DB-WAX 1883. GC/MS m/z (%): 39 (13), 41 (9), 51 (11), 65 (13), 77 (26), 78 (8), 79 (10), 91 (100), 92 (25), 105 (35), 115 (11), 117 (13), 119 (50), 120 (16), 129 (7), 130 (12), 133 (7), 147 (11), 148 (53) [M]⁺.

(e) cis- and trans-2,3,3A,7A-tetrahydro-1H-indene-4-carbaldehyde (5/6): RI_DB-WAX 1924. GC/MS m/z (%): 39 (14), 41 (11), 51 (12), 65 (17), 77 (23), 78 (11), 79 (11), 91 (100), 92 (28), 105 (27), 115 (10), 117 (12), 119 (47), 120 (23), 130 (17), 147 (12), 148 (42) [M]⁺. RI_DB-WAX 1972. GC/MS m/z (%): 39 (11), 41 (8), 51 (8), 65 (13), 77 (14), 78 (6), 79 (12), 91 (100), 92 (28), 105 (11), 115 (14), 117 (16), 119 (38), 120 (22), 130 (16), 147 (12), 148 (41) [M]⁺.